Crystallographic analyses have been successfully initiated for wild type recombinant human Cu,Zn superoxide dismutase (HSOD) and site- directed mutants. Recombinant HSOD as expressed in yeast has been crystallized in four different crystal forms. The three forms characterized to date all diffract to high resolution, are stable to x-rays, and appear suitable for structure determination. Over 30 specific site-directed mutants have been designed, cloned, and expressed, as well as 10/6 more mutants with random amino acids at selected sites. Crystals of a doubly mutated enzyme with increased thermal stability (Cys6 - Ala, Cys111 - Ser), grown de novo and form both micro- and macro-seed of the wild type protein, demonstrate the feasibility of isomorphous crystallization of site- directed mutants of the cloned parent enzyme for comparative structure-function studies. The proposed work will involve 1) the design, construction, expression, testing, and purification of mutant HSOD's, 2) crystallization, data collection, and atomic structure determination for both wild type and mutant enzymes, 3) detailed analysis of these HSOD structures including molecular dynamics calculations together with computational and computer graphic studies, and 4) the recursive application of the results to the rational redesign of HSOD and to the improvement of computational analysis and prediction. Predictive failures will provide an important source of data for the improvement of current modeling technology. The results of the proposed work are intended to probe structural determinants for tertiary folding, quaternary assembly, crystallization, stability, enzyme mechanism, and electrostatic recognition in HSOD. One proposed design principle is the use of structural elements and subdomains as building blocks. Thus, HSOD can also be considered a model for other beta barrel proteins including structural proteins such as crystallin and icosahedral virus coat proteins immunoglobulins electron transport proteins, and several enzymes and binding proteins. Thus, this proposal seeks to understand the structure and function of HSOD at the atomic level and to both test and use this understanding by making specific structural changes to modify the stability and activity of the wild type enzyme in predetermined ways. This knowledge should contribute to the redesign of HSOD for protective drug use against ischemic damage, by allowing predetermined changes in enzyme stability and activity, and by identifying regions where allowed changes could be used for targeting to specific tissues and/or altering the serum lifetime of the enzyme. In the broader sense, the results of the proposed work aim to provide a fundamental basis for rational protein engineering and design that can be applied to many systems of scientific medical, and commercial interest.